An Advanced Skyline Approach for Imperfect Data Exploitation and Analysis

Abstract : The main purpose of this thesis is to study an advanced database tool named the skyline operator in the context of imperfect data modeled by the evidence theory. In this thesis, we first address, on the one hand, the fundamental question of how to extend the dominance relationship to evidential data, and on the other hand, it provides some optimization techniques for improving the efficiency of the evidential skyline. We then introduce efficient approach for querying and processing the evidential skyline over multiple and distributed servers. ln addition, we propose efficient methods to maintain the skyline results in the evidential database context wben a set of objects is inserted or deleted. The idea is to incrementally compute the new skyline, without reconducting an initial operation from the scratch. In the second step, we introduce the top-k skyline query over imperfect data and we develop efficient algorithms its computation. Further more, since the evidential skyline size is often too large to be analyzed, we define the set SKY² to refine the evidential skyline and retrieve the best evidential skyline objects (or the stars). In addition, we develop suitable algorithms based on scalable techniques to efficiently compute the evidential SKY². Extensive experiments were conducted to show the efficiency and the effectiveness of our approaches.
Document type :
Theses
Complete list of metadatas

Cited literature [94 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-01591846
Contributor : Abes Star <>
Submitted on : Friday, September 22, 2017 - 9:45:24 AM
Last modification on : Wednesday, September 5, 2018 - 1:30:09 PM
Long-term archiving on : Saturday, December 23, 2017 - 12:59:02 PM

File

2017ESMA0011_elmi.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01591846, version 1

Collections

Citation

Saïda Elmi. An Advanced Skyline Approach for Imperfect Data Exploitation and Analysis. Other [cs.OH]. ISAE-ENSMA Ecole Nationale Supérieure de Mécanique et d'Aérotechique - Poitiers, 2017. English. ⟨NNT : 2017ESMA0011⟩. ⟨tel-01591846⟩

Share

Metrics

Record views

313

Files downloads

326